mysterymath wrote: > The only real 'gotcha' I can see is that thanks to the sorting, we allow > 'different' attributes, as long as the 'aspects' are all the same (that is, > `float, double` and `double, float`) thanks to the sort. > > I question the value of that sort, and probably would just want to disallow > duplicate attributes unless identical, but if you have a good reason for it, > I'm Ok with it.
It's mostly just annoyance at "overly conservative" error reporting; it's logically a set, so it should behave like a set. There's no real meaning associated with the order, so it would be odd for different orders to produce observable differences in behavior. https://github.com/llvm/llvm-project/pull/147431 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
